P.D. JENKINS AND M.F. ROBINSON 
gymnura 
aureospinula 
truei 
sinensis 
suillus 
hainanensis 
parvus 
megalotis 
Erinaceinae 
E. gymnura 
P. aureospinula 
P. Enwer 
H. sinensis 
H. hainanensis 
ishh Sibi Il Wats) 
lel Sen Vea AUNS) 
H. megalotis 
Erinaceinae 
Fig. 4 Comparison of trees obtained for the Galericinae. (a) One of twenty most parsimonious trees, all showing the same configuration for the 
Galericinae. Tree length 141 steps, with a Consistency Index of 0.72, a Retention Index of 0.93, and a Rescaled Consistency index of 0.66. The 
branching pattern on the left shows the results from the analysis of this study, that on the right is partially redrawn from Frost er al. (1991: fig. 9), 
restricted to show only the relationships within the Galericinae and is 128 steps in length with a Consistency Index of 0.76. (b) One of the remaining 
twenty most parsimonious trees, showing the alternate arrangement for the Galericinae. 
[66.1] M3 hypocone (see Frost. et al. 1991) or metastylar spur (see 
Gould, 1995): (0) absent or weak; (1) present, well developed on 
buccal side. CI 1.000. 
[69.1] Axis, posteroventral keel: (0) absent; (1) present. CI 1.000. 
[71.1] Scapula, metacromium process: (0) deltoid, amorphous pro- 
jection; (1) long, fusiform projection. CI 1.000. 
[72.1] Sacral vertebrae, neural spines: (0) not fused into continuous 
longitudinal plate; (1) fused into continuous longitudinal plate. CI 
1.000. 
[73.1] Ischium, posterodorsal process (see Gould, 1995 for correc- 
tion of error by Frost et al. 1991): (0) not greatly elongated; (1) 
greatly elongated. CI 1.000. 
[74.1] Tibia, lateral flange on antero-superior margin: (0) absent or 
weakly present; (1) strongly developed. CI 1.000. 
SYNAPOMORPHY OF HYLOMYS: 
[19.1] Cranio-orbital foramen in the frontal: (0) closely associated or 
joined with the ethmoid foramen; (1) foramina widely separated. 
The terminology for this character is confusing. Frost er al. (1991) 
used the name ophthalmic foramen (which they attributed to Butler 
(1948) although this name could not be found in this paper), but 
Gould (1995: character 19) pointed out that this foramen had been 
misidentified by Butler and is the anterior opening for the superior 
ramus of the stapedial artery. Gould also referred to this foramen as 
